What are the the issues associated with map and reduce slots based mechanism ?
-
-
What are the the issues associated with map and reduce slots based mechanism in Hadoop version 1?
-
In Hadoop, each MapReduce job is split into the task. Task Tracker run each task on a fixed number of map and reduce slots inside a data node based on a static configuration.
Issues associated with map and reduce slots are as follows:
- In a slot based model, it is hard for non-MapReduce applications to be scheduled appropriately.
- Slots are wasted, anytime the cluster workload does not fit the static configuration.
- Jobs never leave their original slots and cannot be moved to free slots
- You must be logged in to reply to this topic.